U.S. patent number 6,072,854 [Application Number 09/117,498] was granted by the patent office on 2000-06-06 for method and apparatus for x-ray topography of single crystal ingot.
This patent grant is currently assigned to Rigaku Corporation. Invention is credited to Tetsuo Kikuchi, Yoshio Machitani.
United States Patent |
6,072,854 |
Kikuchi , et al. |
June 6, 2000 |
**Please see images for:
( Certificate of Correction ) ** |
Method and apparatus for X-ray topography of single crystal
ingot
Abstract
There is disclosed a method for X-ray topography wherein an
X-ray topographic image is obtained for an unprocessed single
crystal ingot as it is after being produced by Czochralski method
to observe crystal dislocations, thereby finding a boundary between
a dislocation-appearing area and a dislocation-disappearing area. A
scanning stage (34) travels along a direction X to position the
observation area of the silicon ingot (22) at the X-ray irradiation
site. The silicon ingot (22) is .phi.-rotated on its axis so that
the predetermined crystal lattice plane erects vertically. A
traveling table (46) travels along a direction Y to align the
rotation center (49) of an .omega.-rotation table (48) with the
peripheral surface of the ingot (22). The .omega.-rotation and
.phi.-rotation are precisely adjusted while observing diffracted
X-rays with an X-ray television camera (26). Then, an X-ray
recording medium (66) is mounted on the scanning stage (34). X-rays
from an X-ray source (24) are incident on the peripheral surface of
the ingot (22) and the scanning stage (34) is scanned by a given
scanning width while recording diffraction images on the X-ray
recording medium (66), thereby obtaining an X-ray topographic image
and finding a boundary between a dislocation-appearing area and a
dislocation-disappearing area.
Inventors: |
Kikuchi; Tetsuo (Tachikawa,
JP), Machitani; Yoshio (Tokyo, JP) |
Assignee: |
Rigaku Corporation (Tokyo,
JP)
|
Family
ID: |
18310736 |
Appl.
No.: |
09/117,498 |
Filed: |
July 28, 1998 |
PCT
Filed: |
December 03, 1997 |
PCT No.: |
PCT/JP97/04414 |
371
Date: |
July 28, 1998 |
102(e)
Date: |
July 28, 1998 |
Foreign Application Priority Data
|
|
|
|
|
Dec 4, 1996 [JP] |
|
|
8-337658 |
|
Current U.S.
Class: |
378/73; 378/71;
378/79; 378/86 |
Current CPC
Class: |
C30B
15/00 (20130101); C30B 29/06 (20130101); C30B
33/00 (20130101); G01N 23/20008 (20130101) |
Current International
Class: |
C30B
33/00 (20060101); C30B 15/00 (20060101); G01N
23/20 (20060101); G01N 023/207 (); C30B
029/06 () |
Field of
Search: |
;378/73,74,71,82,83,49,86,72,79 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
61-45178 |
|
Oct 1986 |
|
JP |
|
62-116243 |
|
May 1987 |
|
JP |
|
62-242845 |
|
Oct 1987 |
|
JP |
|
3-255948 |
|
Nov 1991 |
|
JP |
|
3-255951 |
|
Nov 1991 |
|
JP |
|
5-203592 |
|
Aug 1993 |
|
JP |
|
WO A1 85/01349 |
|
Mar 1985 |
|
WO |
|
Other References
The Rigaku Journal, vol. 8, No. 2, 1991, pp. 25-28, Rigaku Corp.,
Tokyo, Japan. .
Applied Physics Letters, vol. 47, No. 12, 1985, M. P. Scott, S. S.
Laderman and A. G. Elliott, "Microscopic identification of defects
propagating through the center of silicon and indium-doped liquid
encapsulated Czochralski grown GaAs using X-ray topography", pp.
1280-1282..
|
Primary Examiner: Porta; David P.
Assistant Examiner: Dunn; Drew A.
Attorney, Agent or Firm: Frishauf, Holtz, Goodman, Langer
& Chick, P.C
Claims
What is claimed is:
1. An X-ray topographic method for minding a boundary between a
dislocation-appearing area and a dislocation-disappearing area in
an unprocessed single crystal ingot as it is after being produced
by Czochralski method, comprising the steps of:
(a) supporting the unprocessed single crystal ingot rotatably on
its axis and movably along its axis;
(b) irradiating X-rays to a peripheral surface of the ingot through
an incident slit having a long and narrow opening, and obtaining an
image of diffracted X-rays from the ingot with an X-ray television
camera, followed by setting positional relationship between the
ingot and an X-ray optical system so that an optimum image could be
obtained;
(c) disposing a two-dimensional X-ray recording medium at a
position where the diffracted X-rays can be detected;
(d) irradiating X-rays to the peripheral surface of the ingot
through the incident slit, and scanning the ingot synchronously
with the X-ray recording medium in an axial direction of the ingot
while recording diffracted X-rays from the ingot; and
(e) finding a boundary between a dislocation-appearing area and a
dislocation-disappearing area by observing a topographic image
recorded on the X-ray recording medium.
2. An X-ray topographic method according to claim 1, wherein any
two-dimensional X-ray imaging device for setting is used instead of
the X-ray television camera.
3. An X-ray topographic method according to claim 2, wherein any
two-dimensional X-ray imaging device for observing X-ray
topographic images is used instead of the X-ray recording
medium.
4. An X-ray topographic method according to claim 3, wherein the
two-dimensional X-ray imaging device for setting is used also as
the two-dimensional X-ray imaging device for observing X-ray
topographic images.
5. An X-ray topographic method according to claim 3 wherein the
two-dimensional X-ray imaging device for observing X-ray
topographic images is a camera device capable of obtaining X-ray
images as electronic signals, and output signals of the camera
device is stored into different memory areas in every given
sampling time during scanning the ingot, instead of scanning the
X-ray recording medium synchronously with the ingot.
6. An X-ray topographic apparatus for finding a boundary between a
dislocation-appearing area and a dislocation-disappearing area in
an unprocessed single crystal ingot as it is after being produced
by Czochralski method, comprising:
(a) two parallel rollers for supporting thereon the unprocessed
single crystal ingot rotatably around an axis of the ingot;
(b) a scanning stage having the rollers;
(c) a scanning mechanism for traveling the scanning stage in an
axial direction of the ingot;
(d) an incident X-ray optical system including an X-ray source, a
collimator connected to the X-ray source, and an incident slit
fixed to a tip of the collimator, wherein the incident slit has a
long and narrow opening perpendicular to a traveling direction of
the scanning stage;
(e) an X-ray television camera for obtaining an image of diffracted
X-rays from the ingot for the use of setting of an X-ray optical
system;
(f) a two-dimensional X-ray recording medium for recording the
diffracted X-rays from the ingot for the use of recording of an
X-ray topographic image;
(g) a medium cassette which can support the X-ray recording medium,
wherein the X-ray recording medium is positioned selectively in a
position on a light path of diffracted X-rays and in a position out
of the light path of the diffracted X-rays;
(h) a rotary table on which the incident X-ray optical system and
the X-ray television camera are mounted, the rotary table being
rotatable around a rotation center;
(i) an angle-adjustment device allowing the X-ray television camera
to rotate on the rotary table around the rotation center so as to
adjust an rotary angle; and
(j) a traveling table on which the rotary table is mounted, the
traveling table being movable in a direction perpendicular to the
traveling direction of the scanning stage.
7. An X-ray topographic method according to claim 4 wherein the
two-dimensional X-ray imaging device for observing X-ray
topographic images is a camera device capable of obtaining X-ray
images as electronic signals, and output signals of the camera
device is stored into different memory areas in every given
sampling time during scanning the ingot, instead of scanning the
X-ray recording medium synchronously with the ingot.
Description
TECHNICAL FIELD
The present invention relates to a method and apparatus for X-ray
topography for observing dislocations in a single crystal ingot,
typically a silicon ingot.
BACKGROUND ART
Silicon wafers are cut off from silicon single crystal ingots. It
has been a recent trend to produce the silicon single crystal
ingots by Czochralski method. FIG. 7 shows a front view of an
unprocessed silicon ingot as it is after being produced by
Czochralski method. The ingot is produced by the steps comprising:
(1) dipping a seed crystal 10 (6 to 12 mm in diameter) in silicon
melt; (2) pulling up the seed crystal to form a necking 12 (about 2
to 4 mm in diameter and about 50 to 100 mm in length) to eliminate
lattice defects such as dislocations and processing defects; (3)
forming a shoulder 14 to thicken the crystal to a diameter by about
1% larger than the desired diameter; (4) growing a straight
cylindrical body 16 to a desired length; and (5) forming a conical
tail 18.
The conical tail 18 is formed so that no dislocations due to
thermal distortion are generated in the straight cylindrical body
16. In forming this conical tail 18, the temperature of the melt is
raised a little and the pull-up speed is raised. Many dislocations
appear in this conical tail 18 due to thermal distortion, often
reaching to the straight cylindrical body 16.
A cylindrical useful crystal 20 is cut off from the ingot mentioned
above and silicon wafers are produced by slicing this useful
crystal 20. The cut-off step is that the shoulder 14 and the
conical tail 18 are removed from the silicon ingot, followed by
processing the periphery of the straight cylindrical body 16 to
adjust its diameter to a desired value. Of the straight cylindrical
body 16 (with a length of L1), the region 15 (with a length of L2)
close to the shoulder 14 and the region 19 (with a length L4) close
to the conical tail 18 are removed since dislocations may appear in
these regions due to thermal distortion. Accordingly, the length L3
of the useful crystal 20 becomes shorter than the length L1 of the
straight cylindrical body 16. While it is desirable to make the
length L3 of the useful crystal 20 as long as possible for the
purpose of effectively utilizing the expensive silicon ingot, the
risk to incorporate the dislocation-appearing area into the useful
crystal 20 is also enhanced when the removed regions 15 and 19 are
diminished. Dislocations are liable to appear especially in the
region 19 close to the conical tail 18. The regions 15 and 19 to be
removed have been rather widely selected larger than the estimated
dislocation areas which are based on the empirical estimation. This
selection means that a significant amount of the good quality area
including no dislocations had been discarded.
It would be possible to take out the maximum useful crystal
containing no dislocations provided that the boundary between a
dislocation-appearing area and a dislocation-disappearing area
could be precisely found. Therefore, it is desired to provide a
detection technique that can find the boundary.
There have been known in the art several methods for observing
dislocations in a silicon single crystal: for example, (1) a
wet-etching method applied to disc-shaped test pieces cut-off from
the silicon ingot; and (2) an X-ray topography method applied to
thin plates cut-off from the ingot, using forward-reflection X-ray
topography (with a Lang camera). These methods, however, require
the cut-off test pieces, so that dislocations could not be observed
in the unprocessed silicon ingot as it is.
Dislocations could be observed without cutting off the silicon
ingot by using, in principle, back-reflection X-ray topography.
However, detection of dislocations by the back-reflection X-ray
topography have been applied only for the disk-shaped test pieces
cut-off from the silicon ingot and no X-ray topographic apparatus
are known that can find the boundary between a
dislocation-appearing area and a dislocation-disappearing area in
the unprocessed silicon ingot which is used as a test sample as it
is.
An object of the present invention is to provide a method and
apparatus for X-ray topography that can observe dislocations and
find the boundary between a dislocation-appearing area and a
dislocation-disappearing area in the unprocessed single crystal
ingot, typically silicon ingot, by using the ingot as it is as a
test sample after being produced by Czochralski method.
DISCLOSURE OF INVENTION
The present invention provides an X-ray topographic method for
finding a boundary between a dislocation-appearing area and a
dislocation-disappearing area in an unprocessed single crystal
ingot as it is after being produced by Czochralski method,
comprising the steps of: (a) supporting the unprocessed single
crystal ingot rotatably on its axis and movably along its axis; (b)
irradiating X-rays to a peripheral surface of the ingot through an
incident slit having a long and narrow opening, and obtaining an
image of diffracted X-rays from the ingot with an X-ray television
camera, followed by setting positional relationship between the
ingot and an X-ray optical system so that an optimum image could be
obtained; (c) disposing a two-dimensional X-ray recording medium at
a position where the diffracted X-rays can be detected; (d)
irradiating X-rays to the peripheral surface of the ingot through
the incident slit, and scanning the ingot synchronously with the
X-ray recording medium in an axial direction of the ingot while
recording diffracted X-rays from the ingot; and (e) finding a
boundary between a dislocation-appearing area and a
dislocation-disappearing area by observing a topographic image
recorded on the X-ray recording medium.
The two-dimensional X-ray recording medium may be an X-ray film or
an imaging plate.
The X-ray detecting device may have various modifications. The
X-ray television camera mentioned above may be substituted by any
two-dimensional X-ray imaging device. In this case, a
two-dimensional detecting surface such as the X-ray television
camera is not always necessary. For example, the detecting device
may be a scintillation detector which is moved up and down and
right to left. The X-ray recording medium mentioned above may be
substituted by any two-dimensional X-ray imaging device for
observing an X-ray topographic image: a camera-type X-ray imaging
device can be used for example. Furthermore, the two-dimensional
X-ray imaging device used for setting may be also used for
observing X-ray topographic images.
The two-dimensional X-ray imaging device for X-ray topographic
observation may be a camera device capable of obtaining X-ray
images as electronic signals. In this case, the X-ray topographic
image can be obtained by storing the output signals of the camera
device into different memory areas in every given sampling time
during scanning the ingot, instead of scanning the X-ray recording
medium synchronouly with the ingot. An outline of this electronic
method will be described below. A distance (scanning unit-width)
where the ingot travels along its axis during a given sampling time
is previously determined to allot a different memory area to each
scanning unit-width. Then, diffraction images are recorded in each
different memory for each given sampling time. After completion of
the scanning over a desired scanning area, the signals recorded in
the memory areas are joined together to form a single image,
thereby obtaining an X-ray topographic image. Thus, the stationary
X-ray imaging device can have the same function as scanning of the
two-dimensional X-ray recording medium.
The present invention also provides X-ray topographic apparatus for
finding a boundary between a dislocation-appearing area and a
dislocation-disappearing area in an unprocessed single crystal
ingot as it is after being produced by Czochralski method,
comprising: (a) two parallel rollers for supporting thereon the
unprocessed single crystal ingot rotatably around an axis of the
ingot; (b) a scanning stage having the rollers; (c) a scanning
mechanism for traveling the scanning stage in an axial direction of
the ingot; (d) an incident X-ray optical system including an X-ray
source, a collimator connected to the X-ray source, and an incident
slit fixed to a tip of the collimator, wherein the incident slit
has a long and narrow opening perpendicular to a traveling
direction of the scanning stage; (e) an X-ray television camera for
obtaining an image of diffracted X-rays from the ingot for the use
of setting of an X-ray optical system; (f) a two-dimensional X-ray
recording medium for recording the diffracted X-rays from the ingot
for the use of recording of an X-ray topographic image; (g) a
medium cassette which can support the X-ray recording medium,
wherein the X-ray recording medium is positioned selectively in a
position on a light path of diffracted X-rays and in a position out
of the light path of the diffracted X-rays; (h) a rotary table on
which the incident X-ray optical system and the X-ray television
camera are mounted, the rotary table being rotatable around a
rotation center; (i) an angle-adjustment device allowing the X-ray
television camera to rotate on the rotary table around the rotation
center so as to adjust an rotary angle; and (j) a traveling table
on which the rotary table is mounted, the traveling table being
movable in a direction perpendicular to the traveling direction of
the scanning stage.
The single crystal ingot produced by Czochralski method may be
typically a silicon ingot, but the present invention may be applied
for single crystal ingots of other materials.
According to the present invention, crystal dislocations can be
observed by obtaining an X-ray topographic image of an unprocessed
single crystal ingot as it is after being produced by Czochralski
method, so that the boundary between a dislocation-appearing area
and a dislocation-disappearing area is found. It was therefore made
possible to cut off the single crystal ingot at the boundary,
making it possible to cut off the maximum proportion of the good
quality crystal area where no dislocations appear.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a perspective view showing one embodiment of the X-ray
topographic apparatus according to the present invention.
FIG. 2 is a plan view of the apparatus shown in FIG. 1.
FIG. 3 is a front view of the apparatus shown in FIG. 1.
FIG. 4 is a perspective view of the supporting device for the X-ray
recording medium.
FIG. 5 is a plan view of the apparatus on which the X-ray recording
medium is disposed.
FIGS. 6(A) and (B) show an X-ray topographic image of the silicon
ingot.
FIG. 7 is a front view of an unprocessed silicon ingot.
FIG. 8 is a perspective view of a positioning jig for positioning a
traveling table.
FIG. 9 is a plan view showing the positioning operation of the
traveling table.
FIG. 10 is an enlarged front view of the positioning operation of
the traveling table.
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a perspective view of one embodiment of the X-ray
topographic apparatus according to the present invention while FIG.
2 and FIG. 3 are its plan view and front view. This X-ray
topographic apparatus makes it possible to handle a silicon ingot
as a sample as it is after being produced by Czochralski method
(that is, the ingot is not processed at all). In the basic function
of this apparatus, the silicon ingot 22 can travel along its axis
(direction X) and can rotate around its axis. X-rays from an X-ray
source are incident on the peripheral surface of the ingot 22.
Diffracted X-rays reflected from the ingot are recorded on an X-ray
recording medium 66 (an X-ray film or an imaging plate)
accommodated in a medium cassette 64 (see FIG. 4). The silicon
ingot 22 is scanned synchronouly with the medium cassette 64 in the
direction X. An image of the diffracted X-rays is recorded on the
X-ray recording medium in the cassette during the scanning, so that
an X-ray topographic image is obtained. Since the X-ray diffraction
intensity from the dislocation site in the ingot becomes larger
than the X-ray diffraction intensity from the area of no
dislocations, the dislocation images appear as high-intensity lines
on the topographic image. The X-ray television camera 26 is used
for setting of the X-ray optical system.
This apparatus can measure X-ray topographic images under such an
angular condition that incident X-rays are diffracted at any one of
crystal lattice planes of (331), (224), (440) and (135),
irrespective of the pull-up orientation of the crystal axis of the
silicon ingot. When the pull-up crystal axis is <100>, a
measurement for the crystal lattice plane (115) is possible in
addition to the measurements for the four crystal lattice planes
described above, while a measurement for the crystal lattice plane
(333) is possible when the pull-up crystal axis is <111>.
The construction of this apparatus will be described in more detail
hereinafter. As shown in FIG. 1, a scanning stage support table 30
is fixed on the top face of a frame 28 by means of four supports 31
(see FIG. 3). Two parallel rails 32 are fixed on a scanning stage
support table 30. A scanning stage 34 can travel on the rails 32 in
the direction X. The traveling movement of this scanning stage can
be used for positioning the observed area of the ingot to the X-ray
irradiation site and for scanning the ingot for obtaining a
topographic image. A screw shaft 36 is rotatably supported by the
scanning stage support table 30. A nut is fixed at the bottom face
of this scanning stage 34. The nut engages with the screw shaft 36.
The drive shaft of a scanning motor 38 is connected to the screw
shaft 36. The scanning motor 38 turns to rotate the screw shaft 36,
allowing the scanning stage 34 to travel in the direction X via the
nut mentioned above. A crystal support table 40 is fixed on the top
face of the scanning stage 34. Two rollers (rubber rollers) 42 are
rotatably supported in parallel by the crystal support table 40.
The drive shaft of a .phi.-rotation motor 44 engages with the two
rollers 42 via a gear mechanism. The .phi.-rotation motor 44 turns
to rotate the two rollers 42 synchronously with each other in the
same direction, allowing the silicon ingot 22 mounted thereon to
rotate on its axis. This .phi.-rotation is used for azimuth
adjustment.
On the top face of the frame 28 is mounted a traveling table 46
which can move in the direction Y. On the traveling table 46 is
mounted an .omega.-rotation table 48 which can rotate around its
rotation center 49 (see FIG. 3) extending vertically. On the
.omega.-rotation table 48 is fixed an arm 50 on which an X-ray
source 24 and a collimator 54 are fixed. The X-ray source 24 is a 2
kW sealed-off X-ray tube with a Cu target, with a normal focus
size, and with the point focus in X-ray take-off direction. The
collimator 54 is about 1 meter long with the inside evacuated to
vacuum. The collimator 54 has an inlet window (X-ray source side)
made of beryllium and an outlet window (incident slit side) made of
Kepton foil. An incident slit 56 is fixed at the tip of the
collimator 54. The opening of the incident slit 56 is long and
narrow and extends vertically (direction Z), i.e. the opening is
perpendicular to the traveling direction of the scanning stage 34
(direction X). The opening is 100 mm in height along the direction
Z and 1 mm in width. The opening of the incident slit 56 is
slightly curved as known in the art of X-ray topography (e.g. see
Japanese Examined Patent Publication No. 61-45178). Plural kinds of
the incident slits are prepared depending on the kind of the
crystal lattice planes for which diffracted X-rays will be
observed. Five kinds of the slits are prepared for this embodiment:
four slits for four crystal lattice planes (331), (224), (440) and
(135), and one common slit for crystal lattice planes (333) and
(115).
A camera guide 58 is fixed on the .omega.-rotation table 48. The
camera guide is a circular arc in shape with the center coinciding
with the rotation center 49 (see FIG. 3). A camera stage 60 engages
with the camera guide 58 so as to turn around the rotation center
49 along the camera guide 58. The X-ray television camera 26 is
fixed on the camera stage 60. The X-ray television camera 26
converts X-ray images into visible lights via a fluorescent plate,
and intensifies optically the visible lights with an image
intensifier which is in intimate contact with the fluorescent
plate, and observes the visible lights with a lens-coupling CCD
camera. The imaging area of the X-ray television camera 26 is a
circle with a diameter of about 20 mm. The X-ray television camera
26 in this embodiment is used for setting only of the X-ray optical
system, i.e. the camera 26 is used not for obtaining topographic
images. The topographic images are recorded on an X-ray recording
medium accommodated in a medium cassette mentioned below.
FIG. 4 is a perspective view of the supporting device for the X-ray
recording medium. The five-inch square flat (or two-dimensional)
X-ray recording medium 66 is accommodated in a recording medium
cassette 64, which can be mounted on a bracket 68. A chuck pedestal
70 is fixed to the bottom end of the bracket 68. A permanent magnet
is buried in the bottom face of the chuck pedestal 70 to lock it on
the top face of the scanning stage 34 (see FIG. 1). The X-ray
recording medium 66 may be an X-ray film or an imaging plate (also
referred to as a storage phosphor or a photostimulated
phosphor).
There will be described hereinafter how to use the X-ray
topographic apparatus. Referring to FIG. 2, the X-ray optical
system is angularly adjusted first to satisfy the desired
diffraction condition. In the case where the pull-up axis of the
silicon ingot extends along, for example, a direction <100>
of the crystal, the X-ray optical system is angularly adjusted so
that X-rays can be diffracted at the crystal plane (115) on the
periphery of the silicon ingot. That is, the camera stage 60 (see
FIG. 1) first turns along the camera guide 58 so that the X-ray
television camera 26 is adjusted at the predetermined diffraction
angle 2 .theta. from the incident X-rays. Then, the
.omega.-rotation table 48 turns so that an X-ray incident angle
.omega. is adjusted at a desired angle from the silicon ingot. The
X-ray incident angle .omega. is adjustable in a range of 20 to 50
degrees in this embodiment.
After completion of the angle adjustment, an unprocessed silicon
ingot 22 produced by Czochralski method is mounted on the two
rollers 42. In the case where the level of the silicon ingot axis
(depending on the diameter of the ingot) is not same as the height
of the X-ray optical system, the level can be adjusted by inserting
height-adjusting spacers between the scanning stage 34 and the
crystal support table 40.
Next, the scanning motor 38 turns to move the scanning stage 34 in
the direction X to position the observing area of the ingot (for
example the area near to the tail in the straight cylindrical body)
at the X-ray irradiation site. Inasmuch as the crystal orientation
can be roughly seen from the habit lines (extending axially)
appearing on the periphery of the silicon ingot 22, the
.phi.-rotation motor 44 turns to rotate the silicon ingot on its
axis so that the plane (115) of the silicon ingot 22 erects
approximately vertically by the help of the habit lines (an azimuth
adjustment by .phi.-rotation).
In the next step, the rotation center 49 of the .omega.-rotation
table 48 is positioned at the peripheral surface of the silicon
ingot 22 by moving the traveling table 46 in the direction Y. The
traveling table 46 is locked by a clamp 47 after the positioning.
This positioning work will now be described in more detail
hereinafter. FIG. 8 is a perspective view of a jig 81 for
positioning the traveling table 46. The positioning jig 81 has a
base 82, a support 84 fixed to the base 82, and a pointer 86
horizontally attached near to the top end of the support 84. The
base 82 is used for being in contact against an L-shaped reference
guide 88 fixed on the top face of the .omega.-rotation table 48.
FIG. 9 is a plan view showing the positioning jig 81 which is set
on the .omega.-rotation table 48, and FIG. 10 is its enlarged front
view. As shown in FIG. 10, when the base 82 of the positioning jig
81 is in contact against the reference guide 88 fixed on the top
face of the .omega.-rotation table 48, the tip of the pointer 86 is
so positioned as to come just on the rotation center 49 of the
.omega.-rotation table 48, i.e. the pointer length protruding from
the support 84 is previously so adjusted. The traveling table 46 is
allowed to travel in the direction Y (right to left direction in
the figure) so that the tip of the pointer 86 of the positioning
jig 81 becomes just in contact with the peripheral surface of the
silicon ingot 22, thereby making an alignment between the
peripheral surface of the silicon ingot 22 and the rotation center
49 of the .omega.-rotation table 48.
With the above adjustment, the plane (115) of the silicon ingot and
the X-ray optical system become in the positional relationship
which satisfies approximately the predetermined diffraction
condition. Next, more precise setting for .omega.-rotation and
.phi.-rotation is carried out while
observing the actual diffracted X-rays. The setting will now be
described below. In the condition where no fine adjustment has been
completed for .omega.-rotation and .phi.-rotation, diffracted
X-rays (extending vertically in line and being slightly curved) may
appear in the monitor display of the X-ray television camera but
make no perfect image, i.e. the top and the bottom of of the X-ray
irradiation field may be partly cut off or be in no focus.
Therefore, .omega.-rotation and .phi.-rotation should be
alternately adjusted by every a fine angle so as to obtain a clear
image from the overall X-ray irradiation field.
In the setting, the .omega.-rotation table 48 is adjustable by
every a fine angle within a range of about 5 degrees by means of a
stepping motor. The accuracy of the angle adjustment is 1/1000
degree per one step of the stepping motor. The angle adjustment for
.phi.-rotation is possible by using another stepping motor as
well.
After completion of the setting of the X-ray optical system, the
supporting device for the X-ray recording medium shown in FIG. 4 is
chucked by the magnet on the scanning stage 34 as shown in FIG. 5
(only the X-ray recording medium 66 is shown in the figure). The
distance between the X-ray recording medium 66 and the peripheral
surface of the silicon ingot 22 is about 50 mm. The preparations
for measurement are thus completed. Then, X-ray topographic images
are observed by the steps described below. X-rays from the X-ray
source 24 pass through a long evacuated collimator 54 and incident
slit 56, and are incident on the peripheral surface of the silicon
ingot 22. Diffracted X-rays from the peripheral surface strike the
X-ray recording medium 66, so that a slightly-curved and
vertically-extending diffraction image 72 is recorded on the X-ray
recording medium as shown in FIG. 6(A) (because the incident slit
is slightly curved). Then, the scanning stage 34 is scanned
(traveling in the direction X) by a predetermined scanning width
(tens to 100 mm) while measuring the diffracted X-rays. There is
now recorded as shown in FIG. 6(B) a diffraction image 74
corresponding to the scanning width, or an X-ray topographic image
of the silicon ingot corresponding to the scanning with. The
topographic image recorded on an X-ray film can be visualized by
developing the film. In the case where the image is recorded on an
imaging plate, the image is excited by a laser beam to generate the
fluorescent light with intensity distribution which can be
displayed on a display device or printed on a sheet of paper to
visualize it.
There are shown in FIG. 6(B) a lot of lines 76 with strong
diffraction intensities are observed in the topographic image 74.
The lines 76 correspond to the images of the diffracted X-rays from
dislocations in the silicon ingot crystal (dislocation images). The
dislocation images usually appear in parallel with a given inclined
angle. If the observation area of the silicon ingot is properly
selected, the dislocation images appear at either right or left
side of the topographic image 74 as shown in FIG. 6(B), showing no
dislocation images at the other side as shown. This means that a
proper choice of the observation area makes it possible to detect
the boundary 78 between a dislocation-appearing area and a
dislocation-disappearing area. If the entire topographic image 74
is covered with only any one of a dislocation-appearing area and a
dislocation-disappearing area, the silicon ingot should be shifted
along its axis to obtain other topographic images again until the
boundary 78 is detected.
When the silicon ingot is cut off at the boundary 78 (or at a
position slightly shifted to a dislocation-disappearing side from
the boundary), a useful crystal of good quality can be obtained
with no loss. When a marker 80 is painted on the peripheral surface
of the silicon ingot, the marker 80 is also recorded on the X-ray
topographic image. Therefore, the determination of the position of
the boundary 78 on the periphery of the silicon ingot is based on
this marker 80.
After the boundary between a dislocation-appearing area and a
dislocation-disappearing area have been found by observing the
X-ray topographic image near to the tail in the straight
cylindrical body of the silicon ingot, the scanning stage is moved
to find the other boundary as well near to the shoulder in the
straight cylindrical body, thereby precisely determining two
cutting positions of the silicon ingot.
* * * * *